Age | Commit message (Collapse) | Author | |
---|---|---|---|
2010-03-29 | Add the instruction for the parameter variable declarations of builtin funcs. | Eric Anholt | |
Matches constructor setup, but I'm not really sure why we make the variable twice. | |||
2010-03-29 | Implement dot() builtin. | Eric Anholt | |
2010-03-29 | Implement exp2() and log2(), and make ir_unop_exp and ir_unop_log be base e. | Eric Anholt | |
Making the base e functions IR operations is not a clear win. i965 doesn't support it, it doesn't look like r600 supports it, but r500 does. It should be easily supportable as a lowering pass, though. | |||
2010-03-29 | Implement builtin length() function. | Eric Anholt | |
2010-03-29 | Add sqrt() builtin as an IR operation. | Eric Anholt | |
Following a discussion in #dri-devel, I think this makes more sense than implementing it as RSQ RCP CMP as Mesa did. The i965 has a hardware sqrt that should work, and AMD is suppposed to be able to implement it as RSQ RCP with an alternate floating point mode so that the 0.0 case is handled like we want. | |||
2010-03-29 | Add builtin pow() function. | Eric Anholt | |
2010-03-29 | Implement some binary gentype builtin functions. | Eric Anholt | |
2010-03-29 | Implement additional unary gentype builtins. | Eric Anholt | |
2010-03-29 | Implement the first builtin function: exp(). | Eric Anholt | |